Powerset multi-class cross entropy loss for neural speaker diarization

1 code implementation • 19 Oct 2023 • Alexis Plaquet, Hervé Bredin

Since its introduction in 2019, the whole end-to-end neural diarization (EEND) line of work has been addressing speaker diarization as a frame-wise multi-label classification problem with permutation-invariant training.

Multi-class Classification Multi-Label Classification +2